CN-122022772-A - System and method for evaluating state and service life health of small gate hoist in irrigation area

CN122022772ACN 122022772 ACN122022772 ACN 122022772ACN-122022772-A

Abstract

The invention relates to the technical field of hydraulic engineering, in particular to a system and a method for evaluating the state and service life health of a small gate hoist in a irrigation area. The state monitoring module collects electrical parameters through the comprehensive current controller and combines YOLOv algorithm to identify the physical state of the gate, the service life prediction module adopts LoRA method to finely tune a large model and output health index and residual service life, the fault diagnosis module fuses two-dimensional data to realize omnibearing judgment, the maintenance recommendation module matches a maintenance scheme with a rule engine through a knowledge graph, and the rear end supports instruction comparison and abnormal notification based on a Flask framework. The invention realizes the automation and accurate monitoring of the equipment state and the life prediction, improves the operation and maintenance efficiency, reduces the cost, provides support for digital twin water conservancy and ensures the water supply safety of the irrigation area.

Inventors

Jia Tianshuo
CHEN YUDU
HU XIAOQING
XU WANTING
Zhan Hongli
LI SHANFEI
YANG XIJUN
Zhao Helai
Wang Fanpeng
ZHAO ZHENYU
LI WENKAI
Jiao Ruxia
SONG XUMIN

Assignees

中国电建集团北京勘测设计研究院有限公司

Dates

Publication Date: 20260512
Application Date: 20260129

Claims (11)

1. The system for evaluating the state and life health of the small gate hoist in the irrigation area is characterized by comprising a state monitoring module, a life prediction module, a fault diagnosis module, a maintenance recommendation module and a rear-end processing module; The output end of the state monitoring module is respectively connected with the input end of the life prediction module and the first input end of the fault diagnosis module in a communication way, and is used for synchronously transmitting the collected electric operation data of the hoist and the physical state data of the gate to the life prediction module and the fault diagnosis module; The output end of the life prediction module is respectively connected with the second input end of the fault diagnosis module and the first input end of the maintenance recommendation module in a communication way and is used for outputting the equipment health state evaluation result and the residual service life prediction result to the fault diagnosis module and the maintenance recommendation module; the output end of the fault diagnosis module is in communication connection with the second input end of the maintenance recommendation module and is used for outputting an omnibearing fault judgment result to the maintenance recommendation module; the output end of the maintenance recommendation module is in communication connection with the first input end of the back-end processing module and is used for outputting a maintenance scheme to the back-end processing module; the back-end processing module establishes bidirectional data interaction with the state monitoring module, the service life predicting module, the fault diagnosis module and the maintenance recommending module respectively, and is used for receiving and storing the data of the modules, coordinating the working time sequence of the modules and outputting the final evaluation result.
2. The system for evaluating the state and life health of a small gate hoist in a irrigated area according to claim 1, wherein the state monitoring module comprises a comprehensive current controller and an image acquisition unit; the output end of the integrated current controller and the output end of the image acquisition unit are both in communication connection with the first data receiving end of the back-end processing module, data acquired by the integrated current controller and the image acquisition unit are distributed to the life prediction module and the fault diagnosis module through the back-end processing module, the integrated current controller is used for acquiring electric operation data of the hoist, and the image acquisition unit is used for acquiring physical state image data of the gate.
3. The system for evaluating the state and life health of a small gate hoist in a irrigated area according to claim 2, wherein the integrated current controller is configured with a high-precision sensing module, and the electrical operation data includes a voltage effective value, a harmonic distortion rate and a power factor; the integrated current controller comprises local data storage, a data output end of the integrated current controller is communicated with a first data receiving end of the back-end processing module through a wired or wireless transmission link, collected electric operation data are transmitted to the back-end processing module, and the collected electric operation data are forwarded to the life prediction module through the back-end processing module.
4. The system for evaluating the state and life health of a small gate hoist in a irrigated area according to claim 2, wherein the image acquisition unit is a camera of a gun camera, and the data output end of the image acquisition unit establishes communication with the second data receiving end of the back-end processing module through a 4G/5G or wired transmission link; the rear end processing module is internally provided with a visual recognition algorithm unit, the input end of the visual recognition algorithm unit is connected with an internal data interface of the rear end processing module and is used for receiving physical state image data, performing frame extraction processing and feature analysis, recognizing the opening and closing state, surface defects and structural integrity information of the gate, and the recognition result of the visual recognition algorithm unit is transmitted to the first input end of the fault diagnosis module through the rear end processing module.
5. The system for assessing the state and life health of a small gate hoist in a lavage area according to claim 4, wherein the visual recognition algorithm unit adopts YOLOv algorithm, and the recognition process of YOLOv algorithm comprises: Detecting a boundary frame of the gate body and the door frame, and calculating the intersection ratio of the gate body and the door frame, wherein the calculation formula of the intersection ratio is as follows: Wherein, the As a bounding box of the target area, Is a reference bounding box; detecting a rusted area on the surface of the gate, and counting the ratio of the pixel area of the rusted area to the total pixel area of the gate; the gate corner point is detected and positioned by combining the key points, and the deviation between the opposite side vector included angle and 90 degrees is calculated Or Hausdorff distance, analyzing the geometrical relationship of the gate corner points or the difference between the actual contour of the gate and the standard rectangle, and the deviation The calculation formula of (2) is as follows: and the output result of YOLOv algorithm is used as the gate physical fault judgment basis and is transmitted to a fault diagnosis module.
6. The system for evaluating the state and life health of a small gate hoist in a irrigated area according to claim 1, wherein the life prediction module is internally provided with a large model unit, an input end of the large model unit receives electric operation data transmitted by the state monitoring module, and an output end of the large model unit outputs an equipment health index and a residual service life; The large model unit carries out fine adjustment based on a structured data set paired by features and labels, wherein the features of the structured data set comprise a current total harmonic distortion rate, a voltage total harmonic distortion rate, accumulated start-stop times, accumulated operation duration, a voltage effective value and a power factor, and the labels of the structured data set are equipment health indexes.
7. The system for evaluating the state and life health of a small gate hoist in a irrigated area according to claim 6, wherein the fine tuning of the large model unit adopts a LoRA parameter efficient fine tuning method, and a LoRA adapter is injected through a PEFT library, and a forward propagation formula of the LoRA parameter efficient fine tuning method is as follows: Wherein, the In order to propagate the output in a forward direction, In order to pre-train the weights, Is a low rank matrix; when the electric operation data is preprocessed, a fast Fourier transform algorithm is adopted to calculate the total harmonic distortion rate, and the calculation formula of the total harmonic distortion rate is as follows: To sum the squares of the effective values of the voltages of the harmonics from the 2 nd harmonic to the highest harmonic considered, Is the effective value of the fundamental voltage; and a AdamW optimizer is adopted for training the large model unit, rank and lora-alpha super parameters are adjusted in the training process, an adaptive query/value layer is designated, and the health index and the residual service life output by the trimmed large model unit are transmitted to a fault diagnosis module and a maintenance recommendation module.
8. The system for evaluating the state and life health of a small gate hoist in a pouring area according to claim 1, wherein a fusion logic unit is built in the fault diagnosis module, a first input end of the fusion logic unit receives the recognition result of the vision recognition algorithm unit, a second input end of the fusion logic unit receives the health state evaluation result of the large model unit, the fault source, the fault position and the fault confidence are determined through cross verification of rules and probability fusion logic, and the fault judgment result of the fault diagnosis module is transmitted to the maintenance recommendation module.
9. The system for evaluating the state and life health of a small gate hoist in a filling area according to claim 1, wherein the maintenance recommendation module is internally provided with a knowledge graph unit and a rule engine unit, the knowledge graph unit stores historical fault maintenance data, the input end of the rule engine unit receives the fault determination result of the fault diagnosis module, the health index of the life prediction module and the residual service life respectively, and the maintenance scheme of the maintenance recommendation module is transmitted to the rear-end processing module by matching the historical fault maintenance data with the real-time data to output a maintenance scheme comprising a maintenance operation step, a maintenance priority and a maintenance opportunity.
10. The system for evaluating the state and life health of a small gate hoist in a irrigated area according to claim 1, wherein the back-end processing module is developed based on a Flask framework of Python, acquires gate opening and closing instruction information through an official port, compares the gate opening and closing instruction information with the gate opening and closing state identified by the visual identification algorithm unit, and judges whether an instruction failure exists; When the fault diagnosis module judges that the equipment has faults or the equipment health index is lower than a preset value, the back-end processing module sends notification information to a preset contact person through a telephone or short message API, and meanwhile, the equipment health evaluation report and the maintenance scheme are output.
11. The method for evaluating the state and life health of the small gate hoist in the irrigation area is characterized by comprising the following steps: s1, synchronously acquiring electric operation data of a hoist and physical state data of a gate through acquisition equipment to form two-dimensional monitoring data; S2, performing field fine adjustment on the general large model based on a preset structured data set, receiving the electric operation data obtained in the first step by utilizing the fine-adjusted model, and outputting a device health state assessment result and a residual service life prediction result; S3, fusing the two-dimensional monitoring data with a health state evaluation result and a residual service life prediction result, and carrying out omnibearing judgment on equipment faults; S4, matching and generating a maintenance scheme based on the fault judging result, the health state evaluating result and the residual service life predicting result by combining historical equipment data; s5, receiving the maintenance scheme, coordinating the working time sequence and outputting a final evaluation result.

Description

System and method for evaluating state and service life health of small gate hoist in irrigation area Technical Field The invention belongs to the technical field of hydraulic engineering, and particularly relates to a system and a method for evaluating the state and service life health of a small gate hoist in a irrigated area. Background In irrigation district hydraulic engineering, headstock gear and little gate are as core accuse water equipment, and its running state directly relates to irrigation district water supply safety and irrigation system stability. The traditional small gates in the irrigation area have huge quantity and distributed dispersion, the state monitoring and management of the equipment and the hoist mainly depend on a manual inspection mode, workers need to perform field inspection one by one along the channel, whether the appearance of the equipment has obvious defects such as rust, deformation and the like or not is observed by naked eyes, the operation temperature of the equipment is sensed by means of touch, the tightness of a connecting piece is measured by means of a simple tool such as a wrench and the like, and whether the equipment has faults or not is judged based on personal experience. However, the conventional management mode exposes a plurality of technical limitations in practical application, and it has been difficult to meet the information and intelligent management requirements of the modern irrigation areas: (1) The equipment health status assessment and life prediction capability is lost, and the maintenance mode is passively and inefficiently. The traditional detection means can only preliminarily judge whether the equipment is in an operation state at present, and can not realize the pre-judgment of the residual service life of the equipment and the tracking of the change trend of the health state through data accumulation and systematic analysis. The maintenance work is in double dilemma of excessive maintenance and insufficient maintenance, namely unnecessary waste of manpower and material resources is caused by lack of scientific basis for replacing equipment which can still normally operate in advance, or sudden shutdown of the equipment is caused by failure of prejudgment of potential faults, and the stability of irrigation scheduling in irrigation areas is seriously affected. (2) The manual dependency is high, and the detection efficiency and the detection precision are difficult to guarantee. Because the irrigation area is small in gate number and wide in distribution range, a large amount of manpower is required to be input to carry out periodic inspection in the traditional mode, and the detection process is time-consuming and labor-consuming. Meanwhile, the detection result is easily influenced by subjective and objective conditions such as experience level, responsibility center, environmental factors and the like of the staff, the high-frequency and full-coverage monitoring requirement is difficult to realize, and detection omission and erroneous judgment can be caused by artificial omission, so that the detection efficiency is low and the accuracy cannot be effectively ensured. (3) Lack of digital intelligent management system, information transfer and processing lag. In the traditional management mode, information such as equipment accounts, operation data, fault records and the like is transmitted and stored mainly through paper document records, oral reports and the like, so that obvious hysteresis exists in information sharing, and data deviation is easily caused by manual recording errors. In addition, lack of digital tools such as intelligent APP, remote monitoring platform and the like, staff can not query relevant data of equipment in real time, and a management layer is difficult to master the running state of the equipment remotely, so that timeliness, convenience and synergy of management work are seriously insufficient. (4) The fault finding and response mechanism is imperfect, and the risk prevention and control capability is weak. Due to the limitation of manual inspection periodicity, equipment cannot be timely perceived after faults occur, small faults are easily gradually distinguished into major faults, maintenance cost is greatly increased, and water supply interruption in irrigation areas can be caused by equipment shutdown, so that agricultural production and water consumption of residents are affected. In view of this, the present invention has been made. Disclosure of Invention In order to solve the technical problems in the prior art, the invention provides a system and a method for evaluating the state and service life health of a small gate hoist in a pouring area, and solves the problems that the management mode of the small gate hoist in the pouring area in the prior art depends on manpower, has low efficiency, is difficult to predict service life and has lagged faults. In order to achieve the above purpose, the techni